Power failure information monitoring and early warning processing method and related device

Abstract

The invention discloses a power failure information monitoring and early warning processing method and a related device, and belongs to the technical field of power system operation and maintenance. The method comprises the following steps: acquiring multi-source power failure related data, and preprocessing the multi-source power failure related data; based on the preprocessed multi-source power failure related data, executing two-dimensional power failure risk quantitative judgment; the two-dimensional power failure risk quantitative judgment comprises frequent power failure risk judgment and large-area power failure risk judgment; generating early warning information according to a result of the two-dimensional power failure risk quantitative judgment; and acquiring real-time address information of the grid members, and distributing early warning work orders to the grid members in combination with the early warning information. According to the method, the accuracy and response speed of power failure risk identification are effectively improved, operation and maintenance resource allocation is optimized, the guarantee capability for livelihood sensitive users is enhanced, and the power supply reliability and the service quality are remarkably improved.

Description

Technical Field

[0001] This invention belongs to the field of power system operation and maintenance technology, and relates to a method and related device for monitoring and early warning processing of power outage information. Background Technology

[0002] With the rapid development of society and the economy, the power system, as a critical infrastructure, is directly related to people's livelihood, industrial production, and social order through its safe and stable operation. In recent years, with the advancement of smart grid construction and the deepening of power system reform, users have increasingly higher requirements for power supply reliability and service quality. However, the increasing complexity of the power grid structure, the high proportion of distributed energy integration, and the impact of extreme weather have led to an increasing uncertainty in the frequency and scope of power outages, posing significant challenges to power outage monitoring and emergency response.

[0003] Currently, traditional methods for monitoring and managing power outage information suffer from numerous technical bottlenecks, making it difficult to meet the refined management and control needs of modern power grids. Regarding data collection and integration, existing systems generally suffer from "information silos," with data fragmented across different business systems such as dispatching, distribution, and marketing, resulting in insufficient comprehensiveness and accuracy of power outage information. For example, traditional fault location relies excessively on a single electrical signal data source, and data sharing between different systems is not effectively achieved, causing location errors to frequently exceed 500 meters and fault diagnosis to take more than 30 minutes, severely impacting repair efficiency. Practice at a power supply station shows that, faced with more than 10 independent digital systems, frontline staff need to frequently operate across systems, resulting in repetitive logins and prominent data silos, significantly hindering power outage response speed.

[0004] In terms of risk assessment and early warning mechanisms, existing methods suffer from limitations such as a single dimension and insufficient flexibility. Traditional assessment models often rely on single indicators or empirical judgments, making it difficult to comprehensively consider multiple dimensions of factors, including outage frequency, duration, and impact range. This leads to inaccurate risk level classification and delayed early warnings. Research indicates that existing power grid accident risk management systems generally lack flexibility, making it difficult to adapt to the changing assessment accuracy requirements brought about by power grid development. Furthermore, they are mostly geared towards individual dispatching agencies, lacking integrated management capabilities at the provincial and municipal levels. The massive blackout in the Iberian Peninsula in April 2025 fully exposed the shortcomings of existing systems in risk prediction and protection response. Increased voltage fluctuations were detected before the accident, but due to unreasonable risk assessment threshold settings and a lack of preventative measures, a large-scale blackout affecting 50 million people ultimately occurred.

[0005] In terms of ensuring people's livelihood and providing precise services, the existing system fails to fully consider the differentiated needs of different user groups. Schools, nursing homes, and medical facilities, which are sensitive to people's livelihood, have extremely high requirements for power supply continuity, but traditional monitoring methods lack targeted risk amplification mechanisms and priority handling strategies, making it difficult to guarantee their power safety. Furthermore, the verification mechanism for the authenticity of customer requests is imperfect; false requests or low-quality information can easily interfere with risk assessment and affect the accuracy of resource allocation.

[0006] The work order dispatch and processing processes suffer from inefficiencies and resource mismatches. Traditional manual dispatch relies on experience-based judgment, resulting in an error rate as high as 15%-20%, with rework costs accounting for over 12% of total enterprise maintenance expenditures. The manual process from receiving a work order to completing its dispatch takes an average of over 30 minutes. Urgent and routine work orders are often mixed together, frequently leading to resource mismatches where highly skilled engineers handle low-difficulty tasks, resulting in engineer utilization rates below 60%. Furthermore, the lack of a quantitative evaluation system for work order processing effectiveness hinders the formation of a closed-loop management system of "monitoring-handling-optimization," thus restricting continuous improvement in service quality.

[0007] In responding to large-scale power outages, existing systems lack sufficient quantification of outage coverage ratios at the line and transformer substation levels, failing to effectively differentiate between planned and sudden outages, resulting in unclear control priorities. The protection system suffers from significant "insufficient adaptability." Traditional protection devices, based on a "synchronous generator-dominated" system design, cannot cope with the rapid response characteristics brought about by the high proportion of power electronic equipment connected, easily leading to misjudgments and failures to operate. Furthermore, the lack of a multi-level impact assessment mechanism incorporating topological relationships makes it difficult to accurately quantify the social impact of large-scale power outages.

[0008] As the power system transforms towards a new structure of "high-proportion renewable energy + power electronic equipment," the limitations of traditional power outage monitoring methods in risk prediction, data fusion, and precise service are becoming increasingly apparent. How to construct a monitoring system that integrates multi-source data, establish a multi-dimensional risk quantification model, and achieve a closed loop of intelligent work order dispatch and effect evaluation has become a key technical challenge for improving power supply reliability and customer satisfaction. Against this backdrop, developing a power outage information monitoring method and system that can integrate multi-source data, achieve dual-dimensional risk assessment, accurately serve public needs, and optimize work order processing is of significant practical importance. Summary of the Invention

[0009] The purpose of this invention is to provide a method and related device for monitoring and early warning of power outage information, so as to solve the technical problems of low accuracy and slow response speed of power outage risk identification in the prior art, which affects the reliability of power supply.

[0010] To achieve the above objectives, the present invention employs the following technical solution: In a first aspect, the present invention provides a method for monitoring and early warning processing of power outage information, comprising the following steps: Acquire multi-source power outage related data and preprocess the multi-source power outage related data; Based on the preprocessed multi-source power outage related data, a two-dimensional power outage risk quantification assessment is performed; the two-dimensional power outage risk quantification assessment includes frequent power outage risk assessment and large-area power outage risk assessment. Early warning information is generated based on the results of the dual-dimensional power outage risk quantification assessment. Obtain the real-time address information of grid workers, and issue early warning work orders to grid workers in combination with the early warning information.

[0011] Furthermore, the step of acquiring multi-source power outage related data and preprocessing the multi-source power outage related data specifically includes: Acquire multi-source power outage related data; the multi-source power outage related data includes at least customer power outage records, power outage information, station-line-transformer-customer topology data preset in the business system, power outage assessment request data, and customer call transcript data; the customer power outage records include the start time and end time of each power outage; the station-line-transformer-customer topology data includes the total number of customers on the line. Total number of customers in Taiwan ; Based on the customer ID, link the customer's power outage records with the customer's basic information to calculate the total number of valid power outage records and the average power outage duration for the target customer in the past 60 days. The average power outage duration The calculation formula is:

[0012] in, This represents the total number of valid power outage records. Spatial correlation matching is performed on the topology data of station-line-transformer-customer and the power outage assessment and demand data to obtain the transformer area-customer association information; the effective power outage record is the power outage record after excluding flash outages; the flash outage is defined as a power outage with a duration of less than 5 minutes; Based on the transformer area-customer association information, aggregate power outage data by transformer area and calculate the proportion P of customers experiencing power outages in each transformer area. Transit areas with values ​​greater than or equal to the first preset threshold are marked as key monitoring areas; Semantic authenticity verification is performed on customer call transcripts to construct a formula for scoring the authenticity of customer requests.

[0013] in, Score for the authenticity of the claim; The number of power outage-related keywords matched in the text; Keyword weight; This is the sentiment value of the text, output through a semantic analysis model. Emotional weighting; A request is considered valid if it is greater than or equal to a preset request threshold. If the power outage record is less than the preset threshold, a second verification will be conducted on the customer's power outage record. The second verification method is to compare with other customers in the same area to see if they experienced power outages at the same time, so as to eliminate the interference of false claims on risk assessment.

[0014] Furthermore, the process for determining the risk of frequent power outages specifically includes: A formula for calculating the risk value of frequent power outages is constructed to quantitatively assess the power outage risk of target customers. The formula for calculating the risk value of frequent power outages is as follows:

[0015] in, This represents the risk value for frequent power outages. The number of effective power outages for the target customer in the past 60 days; Weighted by the number of power outages; Average power outage duration, in hours; Weighted by power outage duration; The number of frequent power outage complaints or feedback tickets associated with the target customer within the past two months; As a weighting factor for complaints; When the risk value of frequent power outages If the value is greater than or equal to the second preset threshold, it is judged as a risk of frequent power outages; When the target customers are those with sensitive livelihood needs, The calculation results are corrected using the following formula: = ×(1+ ) in, This is the revised risk value for frequent power outages; A sensitive factor for people's livelihood; When the revised risk value of frequent power outages When the value is greater than or equal to the third preset threshold, it is judged as a risk of frequent power outages; The process for determining the risk of a large-scale power outage specifically includes: A formula for the impact index of large-area power outages is constructed, which combines topological relationships to quantify the impact of regional power outages. The specific calculation formula is as follows:

[0016] in, This is the impact index of a large-scale power outage; α is the line outage weight, and β is the transformer area outage weight. When the transformer area is a key area of ​​concern, the values ​​of α and β are adjusted. This represents the actual number of customers experiencing power outages on the same power line. γ represents the actual number of customers experiencing power outages within the same transformer area; γ is the power outage planning factor. When a large-scale power outage affects the index If the value is greater than or equal to the fourth preset threshold, it is considered a risk of large-scale power outage.

[0017] Furthermore, the process for determining the risk of frequent power outages also includes: Construct a frequent power outage model based on complaints and opinions, and configure data element filtering rules based on specific relevant business attributes in the business system to filter out high-frequency risk work orders; The high-frequency risk work orders are included in the training samples of the frequent power outage model based on complaints and opinions, and the formula for calculating the risk value of frequent power outages is periodically optimized. Weight; The weights are optimized quarterly, and the optimization method is to adjust them using gradient descent based on the matching degree between historical risk events and their actual impact.

[0018] Furthermore, the step of generating early warning information based on the result of the dual-dimensional power outage risk quantification specifically includes: Generate a function based on the results of the dual-dimensional power outage risk quantification assessment. or The specific numerical values ​​and risk level warning information; the risk level is based on... and The specific values ​​are divided into red alert, orange alert, and no alert; The specific formula for calculating the priority of power outage risk warnings is as follows:

[0019] in, Assign a score to the priority of early warnings; The warning information is pushed to local management personnel and sorted by Z-value from high to low. At the same time, the coordinate points of the power outage risk areas are displayed and marked. and Numerical value.

[0020] Furthermore, the step of obtaining the grid worker's real-time address information and issuing early warning work orders to the grid worker in conjunction with the early warning information specifically includes: Obtain real-time address information of grid workers and calculate the straight-line distance between grid workers and power outage risk areas. Prioritize dispatching early warning work orders to Grid workers whose distance is less than or equal to the preset distance threshold and who currently have no outstanding work orders.

[0021] Furthermore, the step of obtaining the grid worker's real-time address information and issuing early warning work orders to the grid worker in conjunction with the early warning information also includes: The effectiveness of handling early warning work orders is quantitatively evaluated, and a work order processing efficiency score is calculated. The specific calculation formula is as follows:

[0022] in, Score based on processing efficiency; T_limit represents the actual processing time of the work order, in hours, calculated from the time the order is dispatched to its completion; T_limit is the work order processing time limit. Processing is considered efficient if the timeliness is greater than or equal to the first preset timeliness threshold. If the time limit is less than the second preset time threshold, it is judged as inefficient processing.

[0023] Secondly, the present invention provides a power outage information monitoring and early warning processing system, comprising: The data processing module is used to acquire multi-source power outage related data and preprocess the multi-source power outage related data; The risk assessment module is used to perform a two-dimensional power outage risk quantification assessment based on preprocessed multi-source power outage related data; the two-dimensional power outage risk quantification assessment includes frequent power outage risk assessment and large-area power outage risk assessment. The early warning module is used to generate early warning information based on the results of the dual-dimensional power outage risk quantification assessment. The dispatch module is used to obtain the real-time address information of grid workers and, in conjunction with the warning information, dispatch warning work orders to the grid workers.

[0024] Thirdly, the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the power outage information monitoring and early warning processing method described above.

[0025] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the power outage information monitoring and early warning processing method described above.

[0026] Compared with the prior art, the present invention has the following beneficial effects: This invention discloses a method and related device for monitoring and early warning processing of power outage information. It acquires multi-source power outage-related data and performs a two-dimensional quantitative assessment of power outage risk, including frequent and large-scale outage risk determination; thereby generating early warning information based on the assessment results. By constructing a standardized analysis process and automated assessment and early warning system based on multi-source data fusion, this invention significantly improves the accuracy and response speed of power outage risk identification, achieving a shift from passive handling to proactive intervention. Through precise dispatching and closed-loop management, it optimizes the allocation of operation and maintenance resources, effectively reducing fault duration, while enhancing the ability to protect sensitive users and improving customer service quality, ultimately comprehensively strengthening the reliability, operational efficiency, and safety management level of the power grid. Attached Figure Description

[0027] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0028] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram of the system of the present invention; Figure 3 This is a system architecture diagram of Marketing 2.0 according to an embodiment of the present invention; Figure 4 This is a business architecture diagram of an embodiment of the present invention; Figure 5 This is an application architecture diagram of an embodiment of the present invention; Figure 6 This is a technical architecture diagram of an embodiment of the present invention; Figure 7 This is a deployment architecture diagram for an embodiment of the present invention. Detailed Implementation

[0029] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other.

[0030] The following detailed description is exemplary and intended to provide further detailed explanation of the invention. Unless otherwise specified, all technical terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terminology used in this invention is for the purpose of describing particular embodiments only and is not intended to limit the scope of exemplary embodiments according to the invention.

[0031] Example 1: See Figure 1 This invention discloses a method for monitoring and early warning of power outage information, comprising the following steps: S1, acquire multi-source power outage related data, and preprocess the multi-source power outage related data; S101, acquire multi-source power outage related data. This multi-source power outage related data includes, but is not limited to, customer power outage records and outage information from customer electricity consumption data collected by the Marketing 2.0 system, the preset "station-line-transformer-customer" topology relationship data in the business system (Marketing 2.0 system), power outage assessment and spontaneous request data from the "Power Grid Map" of the power grid resource business platform, and customer call translation text data from the 95598 business support system. The customer power outage records include the start and end times of each power outage. The "station-line-transformer-customer" topology relationship data includes the total number of customers on the line. Total number of customers in Taiwan ; S102, preprocess the multi-source power outage related data. The preprocessing includes associating customer power outage records with customer basic information based on customer number, and calculating the total number of valid power outage records for the target customer in the past 60 days. Average power outage duration The system spatially correlates and matches spontaneous power outage requests from the "Power Grid Map" with the "Station-Line-Transformer-Customer" topology data to determine the corresponding lines and transformer substations. Valid power outage records are those excluding flash outages, defined as outages lasting less than 5 minutes. The average power outage duration... The calculation method is as follows ; Preferably, the preprocessing in step S102 further includes aggregating power outage data by transformer area by combining the "transformer area-customer" association information of the Marketing 2.0 system, and calculating the proportion of customers experiencing power outages in each transformer area. The calculation method is as follows ;Will Areas with a coverage rate greater than or equal to 30% are marked as "areas of key concern"; calculation is performed in step S2. At that time, the β weight of the key monitoring area was increased to 0.6, and the α weight was reduced to 0.4.

[0032] Preferably, before proceeding with the power outage risk assessment, it is also necessary to perform semantic authenticity verification on the transcribed text data of 95598 customer calls, and construct a formula for calculating the authenticity of the request to assist in verifying the validity of the customer's power outage request. The formula is as follows:

[0033] in, To determine the authenticity of the claim, the score ranges from 0 to 10. The number of keywords in the text that match, including but not limited to, "power outage", "no power", "multiple power outages", and "no power supply". Each keyword is assigned 1.5 points for keyword weight. The value is fixed at 1.5; s is the text sentiment tendency value, output by a semantic analysis model. Negative sentiment corresponds to s values ​​of 2 to 4, and neutral sentiment corresponds to... The value ranges from 0 to 1, with a positive emotion corresponding to a value of 0. The emotional weight is fixed at 1. A value of 6 or higher is considered a valid claim. If the value is less than 6, a second verification will be conducted on the customer's power outage records. The second verification method is to compare with other customers in the same area to see if they experienced power outages during the same period, in order to eliminate the interference of false claims on risk assessment.

[0034] S2, based on the preprocessed multi-source power outage related data, perform a two-dimensional power outage risk quantification judgment; the two-dimensional power outage risk quantification judgment includes frequent power outage risk judgment and large-area power outage risk judgment. S201, Frequent Power Outage Risk Assessment: A formula for calculating the frequent power outage risk value is constructed to quantitatively assess the power outage risk for target customers. The formula is as follows:

[0035] in, This represents the risk value for frequent power outages. The number of effective power outages for the target customer in the past 60 days; The weight for the number of power outages ranges from 0.4 to 0.6. When greater than or equal to 3 Take the upper limit of the value range. When equal to 1 Lower limit of the range of values; Average power outage duration, in hours; The weight for power outage duration is 0.2 to 0.3. When greater than or equal to 2 hours Take the upper limit of the value range. When less than 0.5 hours Lower limit of the range of values; The number of complaints or feedback tickets related to "frequent power outages" associated with the target customer in the past two months; The complaint weight is fixed at 0.2; when there are no complaints... The value is 0; If the value is greater than or equal to 5, it is considered a risk of frequent power outages; Preferably, when the target customers are those with sensitive livelihood needs, then... In addition to the calculation results, a factor sensitive to people's livelihood is added. The fixed value is 0.3, which is the corrected risk value for frequent power outages. The calculation method is as follows = ×(1+ ); after correction A value of 4.5 or higher is considered a risk of frequent power outages; the aforementioned customers with sensitive livelihoods include schools, nursing homes, and users requiring oxygen therapy, which are identified through Marketing 2.0 user profile information.

[0036] S202, Large-scale power outage risk assessment: Constructing a large-scale power outage impact index formula and combining it with topological relationships to quantify the impact of regional power outages, the formula is as follows:

[0037] in, The index represents the impact of a large-scale power outage; α is the line outage weight, with a fixed value of 0.5; β is the transformer substation outage weight, with a fixed value of 0.5. This represents the actual number of customers experiencing power outages on the same power line. The actual number of customers experiencing power outages within the same transformer area; γ is the power outage plan factor, which takes a value of 0.3 when there is a corresponding power outage plan and a value of 1.0 when there is no power outage plan. If the value is greater than or equal to 0.6, it is considered a risk of large-scale power outages; Preferably, the frequent power outage risk assessment in this step also includes constructing a frequent power outage model based on complaints and opinions: This involves configuring data element filtering rules based on the business attributes of "Opinions - Power Supply Quality - Multiple Power Outages" and "Complaints - Power Supply Quality - Power Supply Reliability - Frequent Power Outages" in the Marketing 2.0 system, to meet... Greater than or equal to 1 and Work orders with a value of 3 or higher are automatically marked as "high-frequency risk work orders"; these "high-frequency risk work orders" are included in the model training samples, and the formula for calculating the risk value of frequent power outages is periodically optimized. Weights; weight optimization is performed quarterly, and the optimization method is to adjust the weights based on the matching degree between historical risk events and their actual impact using gradient descent. The accuracy rate of the judgment is greater than or equal to 90%.

[0038] S3, generate early warning information based on the results of the dual-dimensional power outage risk quantification assessment. The warning information carries or The specific values ​​and risk levels will be sent to local management personnel via platform messages and SMS. Simultaneously, the coordinates of the power outage risk areas will be marked and displayed on the "Power Grid Map". and Numerical value; the risk level classification standard is to reach Greater than or equal to 8 and A value greater than or equal to 0.8 triggers a red alert; reaching this level... Greater than or equal to 5 and less than 8 An orange alert is issued if any value between 0.6 and 0.8 is present. Less than 5 and No warning is given when the value is less than 0.6; The preferred option also needs to be based on and A formula for prioritizing power outage risk warnings is constructed to determine the order in which warning information is sent. The formula is as follows: in, The score is the priority score for early warning, with a value ranging from 0 to 9; Normalization is required; the processed value will range from 0 to 10. When pushing data, press... Sort the values ​​from highest to lowest. Warning messages with a value of 7 or higher will be prioritized and pushed to the mobile application of local management personnel, triggering a pop-up reminder.

[0039] S4. Obtain the real-time address information of the grid worker, and issue a warning work order to the grid worker in combination with the warning information.

[0040] The real-time address information of grid workers is obtained, and the latitude and longitude of the grid workers are obtained by connecting to the map system; the straight-line distance between the grid workers and the power outage risk areas is calculated. Prioritize dispatching early warning work orders to For grid workers whose work is within 3 kilometers or less and who currently have no outstanding work orders, accurate work order assignment will be implemented.

[0041] Preferably, the method in this embodiment further includes: S5, quantitatively evaluating the processing effect of the early warning work order.

[0042] The formula for scoring work order processing efficiency is as follows:

[0043] in, The efficiency score ranges from 0 to 100. T_limit represents the actual processing time of the work order, in hours, calculated from dispatch to completion; T_limit represents the work order processing time limit, including work orders at risk of frequent power outages. A value of 4 hours is used for work orders with a risk of large-scale power outages. The value is 2 hours; A value greater than or equal to 80 is considered "high-efficiency processing". A value less than 60 is considered "inefficient processing"; The value is synchronized to the Marketing 2.0 service quality management module for use in grid worker performance evaluation.

[0044] Example 2: This invention discloses a power outage information monitoring and early warning processing system, including a data processing module, a risk assessment module, an early warning module, and a dispatch module. Details are as follows: The data processing module is used to acquire and preprocess data related to multi-source power outages. This module establishes service integration interfaces with the Marketing 2.0 system, the power grid resource business platform, and the 95598 business support system. It acquires customer power outage records, "station-line-transformer-customer" topology data, power outage assessment request data, and call transcript data in real time. The data transmission frequency with the Marketing 2.0 system is 15 minutes per transmission to ensure timely calculations. It includes a built-in data cleaning and parameter calculation unit, which performs customer number association, removes flash outage data, automatically calculates the parameters required for formulas such as the number of effective power outages and the average power outage duration, and stores the calculation results in the corresponding components.

[0045] The risk assessment module is used to perform dual-dimensional power outage risk quantification based on preprocessed multi-source power outage related data. This dual-dimensional risk quantification includes frequent power outage risk assessment and large-area power outage risk assessment. This module incorporates five types of algorithm formulas and parameter configuration units: frequent power outage risk value, large-area power outage impact index, appeal authenticity score, warning priority score, and work order processing efficiency score. It supports users adjusting parameter weights within a reasonable range based on regional power supply characteristics, automatically calculating risk values ​​or indices based on preprocessed parameters, and outputting risk assessment results and warning levels. Preferably, this module includes a historical data training unit, which can import power outage event data from the past 12 months and optimize the risk formula weight parameters through a linear regression model, achieving a risk assessment recall rate ≥92% and a precision rate ≥88%. The system is deployed on the State Grid Cloud Platform, using Kubernetes containers to achieve elastic scaling of the risk quantification module (automatically adding 2 computing nodes when concurrent users ≥200), with storage components configured with 3 8C64G nodes, supporting fast retrieval and visualization based on risk values, with a response time ≤3 seconds.

[0046] The early warning module is used to generate early warning information based on the results of the dual-dimensional power outage risk quantification. This module pushes early warning information according to the early warning priority score.

[0047] The dispatch module is used to obtain the real-time address information of grid workers and, in conjunction with the early warning information, dispatch early warning work orders to them. This module interfaces with the map system to calculate the distance between the grid worker and the risk area, generates dispatch instructions based on a distance of ≤3 kilometers, and pushes them to the grid worker's mobile application.

[0048] Preferably, it also includes an effectiveness evaluation module to quantitatively evaluate the processing effectiveness of early warning work orders and calculate the work order processing efficiency score. This module will calculate the efficiency score and synchronize it to the Marketing 2.0 service quality management module, and generate quarterly weight optimization suggestions based on the score distribution and risk judgment accuracy.

[0049] By adopting the above technical solutions, and constructing a standardized analysis process and automated judgment and early warning system based on multi-source data fusion, the accuracy and response speed of power outage risk identification are significantly improved, realizing a shift from passive handling to proactive intervention. Through precise dispatching and closed-loop management to optimize the allocation of operation and maintenance resources, the duration of faults is effectively reduced, enhancing the ability to guarantee services for users sensitive to public welfare and improving customer service quality. Ultimately, this comprehensively strengthens the reliability of power grid supply, operational efficiency, and safety management level. This is a case study of intelligent analysis and judgment of omnichannel customer service risks based on Marketing 2.0.

[0050] like Figure 3 As shown, the system of this invention is initially integrated with the Marketing 2.0 system, the power grid resource business platform, and the 95598 business support system, and is constructed with multi-source power outage data fusion processing, dual-dimensional risk quantification judgment, and accurate push of early warning information as its functional orientation.

[0051] like Figure 4 As shown, the data flow architecture of the power outage information monitoring system is presented, which mainly includes two parts: internal and external data integration flow and business application data flow. The basic support data flow obtains customer power outage records, "station-line-transformer-customer" topology data, power outage assessment request data, and call transcript data from Marketing 2.0, the power grid resource business platform, and the 95598 system through the data acquisition module; the application data flow mainly targets business application ports such as risk assessment, early warning generation, and work order dispatch.

[0052] See Figure 5 The application architecture of the data system is reflected in the collaboration of various business modules: the data preprocessing module performs data cleaning and transformation processes such as customer number association, temporary power outage exclusion, and parameter calculation; the risk quantification and judgment module has built-in algorithm formulas such as frequent power outage risk value and large-scale power outage impact index to perform core risk calculations; the early warning and dispatch module generates early warning information based on the judgment results and achieves accurate dispatch; and the effect evaluation module completes the evaluation of work order processing efficiency and model optimization.

[0053] See Figure 6This system utilizes service integration interfaces to acquire multi-source power outage-related data, including power outage records, topology data, spontaneous power outage assessment requests, and call transcripts from customer electricity consumption data. Simultaneously, it constructs output channels to support analysis results such as risk assessment, early warning information, and work order data, providing them to data requesters, including local management personnel, through platform in-site messages, SMS, and coordinate point marking on the "Power Grid Map".

[0054] See Figure 7 The technical architecture of the data system is as follows: System data is mainly based on customer power outage records, "station-line-transformer-customer" topology data, power outage assessment request data, and call transcript data. It is acquired and stored through service integration interfaces, and automatically calculated by the built-in algorithm formula unit to output judgment results such as frequent power outage risk value and large-area power outage impact index. The early warning and dispatch module calls the map system interface to obtain grid member real-time location data (latitude and longitude) and calculates the straight-line distance to achieve accurate dispatch within a 3-kilometer range. The analysis results and work order processing efficiency scores are synchronized to the Marketing 2.0 service quality management module through the data interface. In terms of calculation and deployment, the technical architecture of the data system is set as follows: The risk quantification judgment module has a built-in historical data training unit that can import power outage event data from the past 12 months. The weight parameters in the formulas for frequent power outage risk value and large-area power outage impact index are optimized through a linear regression model to ensure that the recall rate of risk judgment is not less than 92% and the precision rate is not less than 88%. This module is deployed on the State Grid Cloud Platform and uses Kubernetes containerization technology for elastic scaling. When the number of concurrent users reaches ≥200, two additional computing nodes are automatically added to ensure performance and reliability in high-concurrency computing scenarios. The data system's storage architecture is configured with three 8C64G nodes to support caching and persistent storage of real-time location data, calculation results, work order data, etc., and supports fast retrieval and visualization based on risk values, ensuring a system response time of ≤3 seconds. All data interactions and API calls comply with power grid information security standards to ensure the security of data transmission and storage.

[0055] Example 3: Implementation of Frequent Power Outage Risk Monitoring and Early Warning – Residential Customer Scenario I. Scene Background A customer in Unit 3, Building 1, Chaoyangyuan Community, Beijing, under the jurisdiction of a State Grid Beijing branch, with customer number BJCY2025001, has called 95598 multiple times in the past two months to report power outages. The local grid staff needs to use the system to determine whether this customer is at risk of frequent power outages to prevent the complaint from spilling over to the 12398 channel.

[0056] II. Input Data (from Marketing 2.0 System and 95598 Business Support System) Customer's power outage records in the past 60 days (excluding flash shutdowns of less than 5 minutes): 1st time (March 10, 2025, 08:15-10:30, duration 2.25 hours), 2nd time (March 25, 2025, 14:40-16:10, duration 1.5 hours), 3rd time (April 8, 2025, 09:05-11:50, duration 2.75 hours). Effective power outage count n=3 times, total power outage duration = 6.5 hours, average power outage duration T≈2.17 hours.

[0057] Complaint ticket data for the past two months: This customer submitted one feedback ticket due to "frequent power outages" (business attribute: "Feedback - Power Supply Quality - Multiple Power Outages"). Therefore, the number of complaints / feedback tickets for the "frequent power outages" category is C=1.

[0058] Weighting parameter (according to system preset rules): Weighting of power outage count =0.6, since n=3 times, take the upper limit of the value range), and the weight of power outage duration. =0.3 (because) ≈2.17 hours ≥2 hours, take the upper limit of the value range), complaint weight. =0.2 (fixed value).

[0059] III. Processing Procedure Data preprocessing: The system uses the Marketing 2.0 "Customer-Power Outage Record" association function to automatically remove one 4-minute intermittent power outage from the customer's three power outage records, calculating the effective number of power outages and the average power outage duration; simultaneously, it synchronizes the customer's complaint ticket data from the 95598 business support system, completing the multi-source data association. This process relies on the reference appendix. Figure 1 The customer management module and metering system management module in the "application layer" enable cross-module linkage between "customer data" and "electricity consumption data" in the data layer, ensuring the accuracy and timeliness of data association.

[0060] The risk value for frequent power outages has been recalculated: due to omissions in the initial data, the effective number of power outages n=6 after supplementation (excluding one flash outage in the last 60 days, with durations of 3 hours, 2.5 hours, 4 hours, 3.5 hours, 2 hours, and 3 hours respectively, for a total duration of 18 hours). =3 hours), number of complaint work orders C=3 times (all work orders in the past 2 months were for the "frequent power outages" category). According to the formula " "calculate: =(6×0.6)+(3×0.3)+(3×0.2)=3.6+0.9+0.6=5.1. The risk value calculation logic is provided in the reference appendix. Figure 4The risk analysis business microservice in the "microservice layer" is deployed under the Spring Boot framework. It calls the calculation results of the data preprocessing module through the Feign component to ensure the automation and accuracy of formula calculation.

[0061] Risk assessment and early warning: due to =5.1≥5, the system determines that the customer has a risk of frequent power outages; combined with the risk level rules (5≤ <8), marked as "Orange Alert". The risk level determination rules are stored in the reference appendix. Figure 3 The "Policy Management" unit of the "Early Warning Management" module allows users to configure thresholds based on regional power supply characteristics. In this judgment, the default policy is invoked, which complies with the standardization requirements of the marketing business domain.

[0062] IV. Output Results Warning Information Generation: The system automatically generates warning information (including "customer number, risk value 5.1, orange warning"), and pushes it to the local management personnel of Chaoyangyuan Community via platform in-system messaging, while simultaneously sending SMS reminders. The information push function relies on the reference appendix. Figure 2 The "Service Dispatch Process" in the "95598 Business Management" domain ensures that early warning information is seamlessly integrated with existing customer service processes, avoiding information silos.

[0063] Visualization: Coordinates are marked on the customer's location within the "Power Grid Map," and the risk of frequent power outages is indicated. =5.1), which facilitates intuitive location for grid workers. This visualization function is provided in the reference appendix. Figure 1 The GIS service module in the "support layer" obtains latitude and longitude coordinates by connecting to the map system, and then synchronizes them to the monitoring theme module in the "application layer" to meet the real-time visualization requirements.

[0064] Precise task assignment triggering: The system obtains the real-time locations of three grid workers in the vicinity, calculates the closest grid worker to the community (within 1.2 kilometers and no outstanding work orders), and automatically assigns the alert work order to their mobile application, requiring them to conduct on-site verification within 4 hours. The task assignment logic is shown in the attached reference. Figure 3 The "Early Warning Management" module is primarily led by the "Comprehensive Management of Early Warning Orders" unit. When this unit connects to the map system to calculate distances, it calls upon the reference appendix. Figure 5 The Redis nodes in the "data storage layer" cache the real-time location data of grid workers, ensuring that the dispatch response time is ≤3 seconds, which meets the system performance requirements.

[0065] Example 4: Assessment and Enhanced Early Warning of Large-Scale Power Outage Risk I. Scene Background The "Tongzhou Development Zone 10kV line (line number: TZKFQ2025-01)" under the jurisdiction of a local branch of State Grid Beijing covers 3 transformer substations. The system needs to monitor whether there is a risk of large-scale power outages on this line to avoid sudden power outages affecting enterprise production and residents' lives.

[0066] II. Input Data (from Marketing 2.0 and Power Grid Resource Business Platform) "Station-Line-Transformer-Customer" Topology Data: Total Number of Customers for Line 10kVTZKFQ2025-01 =2000 households; Total number of customers in "Tongzhou Development Zone A Station Area" under the line =500 households, "B area" =600 households, "C area" =900 households.

[0067] Actual power outage data: Number of customers actually experiencing power outages on the lines =1600 households (covering three transformer substations: A, B, and C); Actual number of customers experiencing power outages in transformer substation A -A=450 households, B area -B=500 households, C area -C=650 households; the power grid resource business platform reported no preset power outage plan, therefore the power outage plan factor... =1.0.

[0068] Weighting parameter: Line outage weight =0.5, power outage weight of transformer area =0.5 (default configuration, non-key monitoring areas).

[0069] III. Processing Procedure Data preprocessing: The system extracts the total number of customers for each line and distribution area from the Marketing 2.0 "Station-Line-Transformer-Customer" module, obtains the real-time number of customers experiencing power outages from the electricity consumption data collection system, and verifies the power outage plan through the power grid resource business platform, completing the spatial correlation and verification of "topology data - power outage data - plan data". This process requires reference to the appendix. Figure 2 The "95598 Power Outage and Restoration Information Reporting Process" within the "Marketing Business Domain" ensures consistency between power outage plan data and business processes; simultaneously, the data verification logic is referenced in the appendix. Figure 4 The Blink component of the "data computation layer" compares multi-source data in real time through stream computing and removes outliers.

[0070] Calculation of the impact index of large-scale power outages: according to the formula " Calculation: Power outage rate = 1600 ÷ 2000 = 0.8; Average power outage rate in the transformer substation = (450÷500+500÷600+650÷900)÷3≈0.818; I = 0.5×0.8+0.5×0.818×1.0=0.809. The index calculation is based on the reference appendix. Figure 5 The "Risk Identification Business Microservice" is executed in the middle. This microservice is deployed on 3 K8S-Worker nodes and uses load balancing (SLB) to distribute computing tasks to ensure computing efficiency in high-concurrency scenarios.

[0071] Risk assessment and escalation warning: Due to I=0.809≥0.6 and no planned power outage. =1.0), the system determines that the line poses a risk of large-scale power outage; based on the risk level rule I≥0.8, it is marked as a "red alert". The risk escalation logic is detailed in the reference appendix. Figure 3 When the "Upgrade Topic Judgment" unit of the "Risk Management" module is triggered, this unit will automatically identify unplanned large-scale power outage scenarios and trigger the upgrade push mechanism.

[0072] IV. Output Results The warning information is escalated and pushed out simultaneously via platform in-site messages and SMS to the local company's operations and maintenance manager and the State Grid Beijing Customer Service Center monitoring console. This also triggers a red marker on the "Power Grid Map" (marked with "Large-area power outage on line 10kVTZKFQ2025-01, I=0.809" and information on the affected transformer areas). See the attached document for the escalation push process. Figure 2 The "Fault Reporting Processing Flow" in the "95598 Business Management" system ensures coordinated response between the operations and maintenance and customer service departments.

[0073] Resource scheduling support: The system retrieves the real-time locations of emergency repair teams near the power line from the grid-based power supply service application, recommends the two closest repair teams (2.3 km and 3.1 km away respectively), and simultaneously provides a list of customers experiencing power outages. The emergency repair team location search relies on the reference appendix. Figure 1 The GIS services of the "support layer" and the structured data storage of the "data layer," along with the inventory synchronization function, are referenced in the appendix. Figure 4 The MQ component implementation in the "microservice layer" ensures the reliability of data transmission through message queues.

[0074] Customer notification: Contact information for customers in the power outage area will be automatically extracted from the Marketing 2.0 user profile, and batch SMS messages with "Power Outage Warning and Estimated Restoration Time" will be sent. Please refer to the attached document for details on customer contact information extraction. Figure 1 Reference for integrating the customer data module and SMS sending function in the "Data Layer" (see attached document). Figure 2 The "Public Information Release Process" in the "Customer Service Management" domain reduces the spillover of customer requests.

[0075] Example 5: Risk Monitoring of Customer Demands Sensitive to People's Livelihoods I. Scene Background The "Haidian Xiyanghong Nursing Home" (customer number: BJHD2025-YLY), which is under the jurisdiction of a local branch of State Grid Beijing, is a customer with sensitive livelihood needs. The system needs to monitor its electricity demand in real time to ensure the electricity supply for the elderly, especially the oxygen equipment.

[0076] II. Input Data 95598 call transcript data (April 15, 2025, 09:30): "Our nursing home has had a power outage since 7 a.m., so the oxygen equipment for the elderly can't be used. We've already reported it once, but it still hasn't been restored. We're extremely worried!" Sensitive Customer Identifier: Extract the "Sensitive Customer - Nursing Home" tag from the Marketing 2.0 user profile, and overlay it with a sensitive factor related to people's livelihood. =0.3.

[0077] Data analysis of the authenticity of the claims: Keyword matching (including "power outage" and "unusable") =2), Emotional Tendency Value ("Extremely Anxious" is a negative emotion, =3) Keyword weight =1.5, Emotional Weight =1.

[0078] III. Processing Procedure Verification of the authenticity of the claim: according to the formula " "calculate: =(2×1.5)+(3×1)=6, because A score of ≥6 is considered a "valid claim." The semantic analysis model, through reference to the appendix... Figure 4 (Technical Architecture Diagram) The DataWorks platform in the "Data Computation Layer" is used for training, and the sentiment values ​​output by the model are synchronized in real time to the "Risk Analysis Business Microservice" to ensure the accuracy of the verification results; at the same time, the appeal verification process must comply with the reference appendix. Figure 3 The "Risk Request Monitoring" module has verification rules that prioritize the authenticity verification of customer requests that are sensitive to people's livelihood issues.

[0079] Frequent power outage risk value correction: Synchronize the power outage records of the nursing home for the past 60 days from the Marketing 2.0 electricity consumption data collection system (after supplementation, n=5 times, T=2 hours, C=2 times), and first calculate the basic risk value: =(5×0.5)+(2×0.3)+(2×0.2)=2.5+0.6+0.4=3.5; Then, add adjustments for factors sensitive to people's livelihoods: =3.5×(1+0.3)=4.55≥4.5, therefore, a risk of frequent power outages is identified. The risk value correction logic is provided in the reference appendix. Figure 5The “Algorithm Analysis Business Microservice” is executed. This microservice has a built-in calculation module for customers with sensitive livelihood issues, which automatically identifies customer tags and triggers correction formulas.

[0080] Risk warning triggered: Based on the sensitivity of public welfare attributes and the revised risk value, the system marks it as a "red warning". The warning level determination should refer to the appendix. Figure 3 The "People's Livelihood Sensitive Monitoring" module has special rules that set lower risk thresholds for clients such as schools and nursing homes to ensure priority response.

[0081] IV. Output Results Warning information will be prioritized and pushed out via a combination of a pop-up window on the mobile application of local management personnel and SMS messages. Simultaneously, the information will be shared with the Haidian Civil Affairs Department's interface, informing them of the power outage's impact and the progress of the response. The priority push mechanism is detailed in the attached reference. Figure 2 The "Special Customer Service Process" within the "Customer Service Management" domain provides support, and the interface with the Civil Affairs Bureau relies on the reference appendix. Figure 4 The Feign component in the "microservice layer" enables cross-departmental data interaction.

[0082] Precise dispatching: The system matches grid workers within 0.8 kilometers of the nursing home with dispatch instructions including "Sensitive customer - Nursing home power outage, priority must be given to ensuring power supply to oxygen equipment," requiring arrival on-site within 2 hours. See attached reference for dispatching priority settings. Figure 3 The "Work Order Priority Rules" in the "Early Warning Management" module set the highest priority for work orders from customers with sensitive livelihood needs; the location calculation of grid workers relies on reference appendices. Figure 1 The GIS service of the "support layer" ensures that the distance calculation error is ≤100 meters.

[0083] Closed-loop tracking: The system updates the work order status in real time. After the grid worker arrives and reports "Line fault confirmed on site, repair team contacted, power expected to be restored within 1 hour," the system synchronizes the progress with the nursing home and civil affairs department until the work order is completed. Final restoration time: April 15th, 11:20 AM, taking 1 hour and 50 minutes. Work order closed-loop management must comply with the reference appendix. Figure 2 The "95598 Quality Evaluation Process" synchronizes processing efficiency scores to the Marketing 2.0 service quality management module for grid worker performance evaluation; the scoring calculation logic is provided in the reference appendix. Figure 5 The "Effect Evaluation Module" is executed. This module is deployed on the node where the RDS database is located to ensure the persistent storage of the score data.

[0084] In one embodiment of the present invention, a computer device is provided, comprising a processor and a memory. The memory stores a computer program, which includes program instructions. The processor executes the program instructions stored in the computer storage medium. The processor may be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing and control core of the terminal, suitable for implementing one or more instructions, specifically suitable for loading and executing one or more instructions in the computer storage medium to achieve a corresponding method flow or corresponding function. The processor described in this embodiment of the present invention can be used in the operation of a power outage information monitoring and early warning processing method.

[0085] This invention also provides a storage medium, specifically a computer-readable storage medium (Memory), which is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the power outage information monitoring and early warning processing method in the above embodiments.

[0086] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0087] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0088] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0089] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0090] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A method for monitoring and early warning processing of power outage information, characterized in that, Includes the following steps: Acquire multi-source power outage related data and preprocess the multi-source power outage related data; Based on the preprocessed multi-source power outage related data, a two-dimensional power outage risk quantification assessment is performed. The dual-dimensional power outage risk quantification assessment includes frequent power outage risk assessment and large-scale power outage risk assessment. Early warning information is generated based on the results of the dual-dimensional power outage risk quantification assessment. Obtain the real-time address information of grid workers, and issue early warning work orders to grid workers in combination with the early warning information.

2. The method for monitoring and early warning processing power outage information according to claim 1, characterized in that, The steps of acquiring multi-source power outage related data and preprocessing the multi-source power outage related data specifically include: Acquire multi-source power outage related data; the multi-source power outage related data includes at least customer power outage records, power outage information, station-line-transformer-customer topology data preset in the business system, power outage assessment request data, and customer call transcript data; the customer power outage records include the start time and end time of each power outage; the station-line-transformer-customer topology data includes the total number of customers on the line. Total number of customers in Taiwan ; Based on the customer ID, link the customer's power outage records with the customer's basic information to calculate the total number of valid power outage records and the average power outage duration for the target customer in the past 60 days. The average power outage duration The calculation formula is: in, This represents the total number of valid power outage records. Spatial correlation matching is performed on the topology data of station-line-transformer-customer and the power outage assessment and demand data to obtain the transformer area-customer association information; the effective power outage record is the power outage record after excluding flash outages; the flash outage is defined as a power outage with a duration of less than 5 minutes; Based on the transformer area-customer association information, aggregate power outage data by transformer area and calculate the proportion P of customers experiencing power outages in each transformer area. Transit areas with values ​​greater than or equal to the first preset threshold are marked as key monitoring areas; Semantic authenticity verification is performed on customer call transcripts to construct a formula for scoring the authenticity of customer requests. in, Score for the authenticity of the claim; The number of power outage-related keywords matched in the text; Keyword weight; This is the sentiment value of the text, output through a semantic analysis model. Emotional weighting; A request is considered valid if it is greater than or equal to a preset request threshold. If the power outage record is less than the preset threshold, a second verification will be conducted on the customer's power outage record. The second verification method is to compare with other customers in the same area to see if they experienced power outages at the same time, so as to eliminate the interference of false claims on risk assessment.

3. The method for monitoring and early warning processing power outage information according to claim 1, characterized in that, The process for determining the risk of frequent power outages specifically includes: A formula for calculating the risk value of frequent power outages is constructed to quantitatively assess the power outage risk of target customers. The formula for calculating the risk value of frequent power outages is as follows: in, This represents the risk value for frequent power outages. The number of effective power outages for the target customer in the past 60 days; Weighted by the number of power outages; Average power outage duration, in hours; Weighted by power outage duration; The number of frequent power outage complaints or feedback tickets associated with the target customer within the past two months; As a weighting factor for complaints; When the risk value of frequent power outages If the value is greater than or equal to the second preset threshold, it is judged as a risk of frequent power outages; When the target customers are those with sensitive livelihood needs, The calculation results are corrected using the following formula: = ×(1+ ) in, This is the revised risk value for frequent power outages; A sensitive factor for people's livelihood; When the revised risk value of frequent power outages When the value is greater than or equal to the third preset threshold, it is judged as a risk of frequent power outages; The process for determining the risk of a large-scale power outage specifically includes: A formula for the impact index of large-area power outages is constructed, which combines topological relationships to quantify the impact of regional power outages. The specific calculation formula is as follows: in, This is the impact index of a large-scale power outage; α is the line outage weight, and β is the transformer area outage weight. When the transformer area is a key area of ​​concern, the values ​​of α and β are adjusted. This represents the actual number of customers experiencing power outages on the same power line. γ represents the actual number of customers experiencing power outages within the same transformer area; γ is the power outage planning factor. When a large-scale power outage affects the index If the value is greater than or equal to the fourth preset threshold, it is considered a risk of large-scale power outage.

4. The method for monitoring and early warning processing power outage information according to claim 3, characterized in that, The process for determining the risk of frequent power outages also includes: Construct a frequent power outage model based on complaints and opinions, and configure data element filtering rules based on specific relevant business attributes in the business system to filter out high-frequency risk work orders; The high-frequency risk work orders are included in the training samples of the frequent power outage model based on complaints and opinions, and the formula for calculating the risk value of frequent power outages is periodically optimized. Weight; The weights are optimized quarterly, and the optimization method is to adjust them using gradient descent based on the matching degree between historical risk events and their actual impact.

5. The method for monitoring and early warning processing power outage information according to claim 1, characterized in that, The step of generating early warning information based on the results of the dual-dimensional power outage risk quantification specifically includes: Generate a function based on the results of the dual-dimensional power outage risk quantification assessment. or The specific numerical values ​​and risk level warning information; the risk level is based on... and The specific values ​​are divided into red alert, orange alert, and no alert; The specific formula for calculating the priority of power outage risk warnings is as follows: in, Assign a score to the priority of early warnings; The warning information is pushed to local management personnel and sorted by Z-value from high to low. At the same time, the coordinate points of the power outage risk areas are displayed and marked. and Numerical value.

6. The method for monitoring and early warning processing power outage information according to claim 1, characterized in that, The step of obtaining the grid worker's real-time address information and issuing early warning work orders to the grid worker in conjunction with the early warning information specifically includes: Obtain real-time address information of grid workers and calculate the straight-line distance between grid workers and power outage risk areas. Prioritize dispatching early warning work orders to Grid workers whose distance is less than or equal to the preset distance threshold and who currently have no outstanding work orders.

7. The method for monitoring and early warning processing power outage information according to claim 1, characterized in that, The step of obtaining the grid worker's real-time address information and issuing early warning work orders to the grid worker in conjunction with the early warning information further includes: The effectiveness of handling early warning work orders is quantitatively evaluated, and a work order processing efficiency score is calculated. The specific calculation formula is as follows: in, Score based on processing efficiency; T_limit represents the actual processing time of the work order, in hours, calculated from the time the order is dispatched to its completion; T_limit is the work order processing time limit. Processing is considered efficient if the timeliness is greater than or equal to the first preset timeliness threshold. If the time limit is less than the second preset time threshold, it is judged as inefficient processing.

8. A power outage information monitoring and early warning processing system, characterized in that, include: The data processing module is used to acquire multi-source power outage related data and preprocess the multi-source power outage related data; The risk assessment module is used to perform a two-dimensional quantitative assessment of power outage risks based on preprocessed multi-source power outage-related data. The dual-dimensional power outage risk quantification assessment includes frequent power outage risk assessment and large-scale power outage risk assessment. The early warning module is used to generate early warning information based on the results of the dual-dimensional power outage risk quantification assessment. The dispatch module is used to obtain the real-time address information of grid workers and, in conjunction with the warning information, dispatch warning work orders to the grid workers.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the power outage information monitoring and early warning processing method as described in any one of claims 1-7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the power outage information monitoring and early warning processing method as described in any one of claims 1-7.

Images